jacobgorm

Unfortunately it just segfaults now: Here is the backtrace from gdb: (gdb) bt #0 __GI___pthread_mutex_lock (mutex=0x8080808080808080) at pthread_mutex_lock.c:65 #1 0x0000007fb7710760 in NvOsMutexLock () from /usr/lib/aarch64-linux-gnu/tegra/libnvos.so #2 0x0000007fab0a88bc in NvMMQueueEnQ ()...

The patch provided here https://github.com/jocover/jetson-ffmpeg/commit/e892195797ada3599a737265150559fbac8d0151 seems to make things worse, I would suggesting reverting it from master until a working solution can be found.

hi again. Are you guys still considering adding half-precision support to BLIS? FWIW there does seem to be a bit of a hole in the market for portable LA library...

Clang now has experimental _Float16 support, but only on ARM : https://clang.llvm.org/docs/LanguageExtensions.html .

@jeffhammond the advantage as the library developer to having _Float16 in the compiler is that it does not promote to float, which should make initial development easier. I agree that...

I agree, this should be two steps, because - in docker builds, it is very slow to debug bundling problems, because you cannot cache the result of the rest of...

One caveat with all the approaches above is that they will expose your Azure secrets to rogue code in any build.rs script you or your dependencies happen to drag via...